Audiences Don’t Suspend Disbelief – We Do It for Them

Our job is to give our audience a great experience. Unfortunately, disbelief can ruin our story’s most important moments. When Gollum sinks into the lava at the end of The Lord of the Rings, audiences are supposed to feel pity, sadness, and relief. They’re not supposed to scoff that this is unrealistic because real lava is too dense for someone to sink into.

Maintaining belief is particularly important to us speculative fiction storytellers, because we ask our audiences to believe in things that are blatantly unrealistic. But belief is always tricky to manage because it is so complex and individual.

In this update to my 2013 article Maintaining Belief During Fantastical Stories (PDF), let’s look at how belief works in fiction and how we can help audiences buy our story events.

The Problem With “Suspension of Disbelief”

While storytellers have naturally discussed belief for millennia, the specific concept of “suspension of disbelief” comes from Samuel Taylor Coleridge’s 1817 Biographia Literaria. At the time, Coleridge was trying to justify his use of the supernatural in poetry by stating that if the poetry had valuable qualities, it could entice readers to use their imaginations to accept the supernatural as real.

However, since then the concept has been broadly applied to suggest that it is up to the audience to believe whatever is on the page. But if audiences could will themselves to believe every story event, why wouldn’t they do that? It would allow them to enjoy stories more. And if audiences could will themselves to believe, why would they accept some story conceits and not others?

There is no doubt that, to some degree, we choose to pretend speculative elements are real because we enjoy them. However, focusing only on this willful intention is ignoring the obvious: that there is a subconscious or emotional component of belief that is not voluntary. That’s why when we violate expectations about what’s real, audiences get a knee-jerk “that doesn’t make sense” reaction. Even if they manage to put that reaction aside, the moment is still ruined.

Besides, we don’t want audiences to have to concentrate and discipline themselves to enjoy our stories. We want them to stay absorbed in our world without thinking about whether they are absorbed.

When audiences experience breaks in belief, it may seem arbitrary or nonsensical to us. Take this quote from an old blog post by John Scalzi. He recalled this story after complaining about someone’s criticism of the lava density in The Lord of the Rings.

When my daughter was much younger, my wife was reading to her from a picture book about a snowman who came to life and befriended a young boy, and on each page they would do a particular activity: build a snow fort, slide down a hill, enjoy a bowl of soup and so on. The last three pages had the snowman walking, then running, and then flying. At which point my wife got an unhappy look on her face and said ‘A flying snowman? That’s just ridiculous!’

To which I said: ‘So you can accept a snowman eating hot soup, but not flying?’ Because, you know, if you can accept the former (not to mention the entire initial premise of a snowman coming to life), I’m not sure how the snowman flying became qualitatively more ridiculous.

Unfortunately, Scalzi is doing some garden-variety audience blaming in this article. He is treating his wife’s reaction as though it is voluntary, and he also demands his wife and the LotR reviewer explain why they find flying snowmen or watery lava unbelievable but not anything else.

While getting frustrated with audience reactions is understandable, we have to accept that reactions are what they are. Our emotions aren’t invalid just because our ability to explain them is spotty at best. Even if the audience disbelieves something because they are factually incorrect about how it works in reality, does that really matter? Their experience still isn’t what we want it to be.

Of course, fixing all moments of disbelief from every single audience member is almost certainly impossible. Don’t burn yourself out attempting it. Still, let’s not encourage ourselves to dismiss problems when we can fix them. After all, these reactions to watery lava and flying snowmen aren’t particularly difficult to explain. They are also both fixable, provided the storytellers wanted to fix them.

How Belief Works, and Why It Breaks

Instead of using the suspension of disbelief as our primary model of believability, Tolkien himself had a better model. He posited something he called “secondary belief.” In brief, this means audiences have a secondary belief system created by credibility and internal consistency within a work. That last part is too reductive to be accurate, but nonetheless it’s much closer than the idea that audiences just will themselves to believe.

Let’s review what we know about audience belief.

• What feels believable is heavily dependent on the story at hand. A giant spider works in Tolkien, but it couldn’t just skitter into the much harder science fiction series The Expanse without ado. This fits Tolkien’s idea that belief is created by internal consistency within each work.
• It’s also readily apparent that speculative fiction tropes become more believable the more an audience gets used to them. Many tropes can be hard for people to swallow when first seen. My first time reading LitRPG, a genre in which fantasy settings operate according to RPG rules, was very strange. But once you read enough of it, things like the characters spontaneously leveling up and getting new skills starts to blend into the background.
• Many believability breaks have nothing to do with the fantastical. This includes Gollum sinking into the lava, a character who somehow falls faster than another character, and a character who gets stabbed through the midsection only to walk around like nothing happened.* All of these unrealistic scenarios are likely to unfold the same way in a nonspeculative work.

Given all of this, we might say that every audience member has an internal model of how the story world works in their head – Tolkien’s secondary belief – but their model pulls from numerous sources. It’s likely to include their ideas about how the real world works, tropes they have seen in similar stories, and whatever they’ve encountered in the story so far. Any one of these sources can create unwritten expectations for the story.

When audiences come up with a mental model of how the story world works, they can extrapolate based on what we show them, but they can’t read our minds. The difference is that extrapolation is relatively specific. If a snowman starts walking and playing alongside a boy, audiences probably expect it to gain other capabilities that make it more boy-like. They’re less likely to expect it to fly, because boys cannot fly.

Similarly, the presence of giant spiders means that giant scorpions should fit right in, and giant animals of other kinds might show up. It doesn’t mean that lava is watery. Those two things have nothing to do with each other, so the presence of giant spiders won’t prevent the lava from breaking believability.

Audiences don’t like unscheduled changes to their model of their world. They want to know important details about how the world works before the story makes use of them. That’s because until the story uses them, they don’t really matter. Unfortunately, storytellers can’t possibly account for all the expectations audiences bring to the story from other places. But we can work to iron out believability breaks that appear for many audience members rather than just a few.

One final addition to our model of believability is the well-known “rule of cool.” This means audiences are much more likely to accept otherwise unbelievable story elements if they think they’re cool – or novel, you might say. I suspect the rule of cool works because it also operates subconsciously. If audiences get an instantaneous “cool!” reaction, it can override an otherwise negative one. However, I don’t recommend relying on this. A story element’s cool factor is also likely to be damaged by disbelief.

How We Set the Rules of Belief

Since an ounce of prevention is worth a pound of cure, we want to work proactively to avoid believability issues. That starts with creating worlds that make it easier for audiences to believe. This is where creating a theme for your world comes in handy. If you make intentional choices about things that are central to believability, you’re more likely to be consistent. That consistency prevents unpleasant surprises.

In speculative fiction, genre is probably the biggest choice we make because it usually determines the type of setting we use. While it’s fun to bend genre conventions, sticking to a genre does help with believability. A genre contains lots of tropes that are associated with each other. These associations make them feel consistent even if there are logical contradictions.

Most of your audience will also be familiar with the story elements typically present in works of the genre. This means audiences are more likely to include those things in their conception of your world to start with, so you don’t have to carefully introduce them to make them believable.

The story’s level of realism also makes a huge difference to which elements feel believable. This usually correlates with genre and tone, but it’s an independent factor. High realism creates a gritty or grounded atmosphere, making story elements feel more real but also creating a stricter set of expectations for what fits in. In contrast, lower realism creates a more whimsical, surreal, zany, or campy feel. Some tropes only feel believable with low realism.

To get a sense of just how low realism can go, let’s look at The Darkangel Trilogy by Meredith Ann Pierce. In this book series, the following happens:

• The protagonist weaves garments out of emotions such as pity, patience, or love for some wraiths to wear.
• The protagonist makes a spontaneous effort to save her love interest by cutting out his lead-covered heart and replacing it with her own. She has no medical training.
• The protagonist questions whether the love interest loves her freely, since he has her heart rather than his own.

In the context of the story, all of these low-realism details fit. They follow the conventions of fairy tales, which are typically very low in realism. In fact, in the Darkangel series it feels less believable that the protagonist lives on a terraformed moon, because this science fiction trope clashes with all of the fairy tale elements of the story. It’s hard to reconcile the more rigorous science of terraforming with the idea that the protagonist can do a heart transplant on a whim or weave garments from emotions.

When we employ elements that seem contradictory, we can try reconciling them for the audience. Let’s say we want computers in our world, but we also want characters to wield swords instead of more advanced weapons. Perhaps we can divide our world between high-resource zones, with lots of technology including computers, and low-resource zones, where guns aren’t even available. Alternately, we might also say guns are banned by a world power or that the world has been so peaceful for so long that people haven’t been inventing weapons.

If our reasoning for how these elements coexist resonates with the audience, the contradiction can make the world more interesting. Unfortunately, if the audience doesn’t buy our argument, it only makes the believability problem worse. Instead of just swallowing the contradiction, the audience also has to swallow our nonsensical explanation. If we simply don’t mention the issue, they stop thinking about it sooner.

This is a big reason why we shouldn’t pile on lots of explanations. If you decide to risk using contradictory tropes, use your best argument and then leave it be. Allow audiences who don’t like your explanation to forget you made it. If a compelling argument is out of reach, don’t make one at all.

While the rules of our world are important, many believability issues don’t happen at the world level. When we establish the characteristics of people, places, and items, it’s always possible for those characteristics to contradict what we say later. This is particularly likely when we avoid pinning down the size and scale of story elements. Figuring out these logistics is often a hassle, but it can prevent issues later. If we don’t know how big our dragons are, it’s easy to imply they are the size of a house in one scene and the size of a horse in another.

When we want to make changes to the audience’s idea of the story world without breaking believability, we have a couple options. First, we can foreshadow by inserting hints about the change. Hopefully when the change appears, it clicks into place with the other hints. Second, we can makes changes gradually. This helps audience extrapolate trends, such as a snowman becoming more boy-like.

---

Unless we outright lie to our audience, which I don’t recommend, they’ll know everything in our story is made up. Unfortunately, that means our creations are under extra scrutiny. To go over smoothly, our story has to feel more believable than the real world. But as long as we’re consistent about how our secondary reality works, it doesn’t have to be remotely possible. For that, we can be grateful.